For safety as well as solubility reasons the Grignard
reaction was carried out in THF rather than in diethyl ether
as reported earlier.10 Because of the poor solubility of
aldehyde 2 in THF the use of toluene as a cosolvent was
found to be important.11 In the event, a fine dispersion of
aldehyde 2 in toluene-THF was reacted with 2.6 equiv12 of
EtMgBr to afford pure 3 in 80% yield after recrystallization.
Small-scale experiments indicated that maintaining the
reaction temperature around 20 °C during the addition of
EtMgBr solution was optimal. At temperatures below 15 °C
the solubility of the forming magnesium phenolate/alcoholate
decreases making stirring difficult. At temperatures higher
than 25 °C coloration, even charring, of the reaction mixture
occurs, decreasing the yield and purity of the product.
Recrystallization of the crude product from a minimum
amount of EtOAc provided pure phenolic alcohol 3 free from
any starting material in good yield. Unless these precautions
(efficient stirring and maintaining the reaction temperature
at 20 ( 5 °C) are taken, the product could contain up to 5%
of the starting aldehyde 2 that, when carried over to the
hydrogenation step, affords m-cresol, which could contami-
nate the final product. Since m-cresol is also behaviorally
active for certain tsetse fly species the final product must be
free from this homologue.
Finally, hydrogenolysis of 3 in methanol at atmospheric
pressure using Pd-on-carbon catalyst gave 1 in nearly
quantitative isolated yield. With smaller batches (<50 g) the
reduction was typically performed at ambient temperature
in ethanol with or without acid catalyst, but on a large scale
it was preferably carried out in methanol in the presence of
70% aqueous HClO4 (ca. 0.03% with regard to solvent) and
at 40 °C with efficient magnetic stirring. Although acetic
acid (up to 10% with regard to solvent) was also found to
facilitate the reduction, its removal, for example by distil-
lation or extraction, complicates workup.
is applicable to the synthesis of other alkylated aromatics if
the corresponding aldehyde is readily available (see, for
example, ref 3).
Experimental Section
Proton and 13C NMR spectra were recorded in CDCl3 at
400 and 100 MHz, respectively, on a Varian spectrometer.
Chemical shifts are expressed in ppm using the solvent signal
1
(CDCl3; δ ) 7.26 for H and δ ) 77.0 for 13C spectra,
respectively) as internal reference. IR spectra were recorded
on a Nicolet Magna-IR 750 spectrometer. Mass spectrometry
was performed on a VG ZAB 2SEQ mass spectrometer in
electron ionization mode. HPLC was performed on an ISCO
2350 system with UV detection at 220 nm through a Hypersil
BDS C18 column (4.6 mm × 150 mm) using a 40:60 mixture
of 0.05 M aqueous KH2PO4 buffer (pH ) 3.5)-methanol
as eluent (1 mL/min). Thin-layer chromatography used 0.25-
mm thick silica gel plates (DC Alufolien Kieselgel 60, Merck
KGaA, Darmstadt, Germany). The Pd-catalyst was from
Merck, other reagents were purchased from Aldrich or Fluka,
while solvents were from Reanal (Budapest, Hungary).
(()-3-(1-Hydroxypropyl)phenol (3). Finely ground 3-hy-
droxybenzaldehyde (2, 1250 g, 10.2 mol) was dissolved in
warm anhydrous toluene (2.2 L). The solution was then
allowed to cool to ca. 30 °C, purged with dry argon gas and
diluted with anhydrous THF (20 L) while stirring using
mechanical stirrer. The effectively stirred suspension was
then cooled to 10 °C, and a solution of EtMgBr, freshly
prepared from ethyl bromide (1987 mL, 26.6 mol) and
magnesium (648 g, 26.6 mol) in anhydrous THF (8.2 L),
was added13 over the course of 3 h while carefully maintain-
ing the reaction temperature between 15 and 25 °C using
water + dry ice as cooling bath. The thick reaction mixture
was then stirred and refluxed for 2 h, cooled to 5 °C,
quenched with cold water (1.0 L), and acidified with 5 M
HCl solution (5.6 L). The phases were separated, and the
aqueous layer was extracted with methyl tert-butyl ether
(4 × 1.0 L).14 The organic phases were combined, washed
successively with water, saturated NaHCO3 solution, and
water (1.0-1.0 L), and dried (MgSO4). The solvent was
evaporated to give a thick oil (ca. 1600 g) that was briefly
stirred with EtOAc (ca. 1.0 L) at 30 °C and then allowed to
crystallize at 5 °C in a refrigerator over 14 h. The product
(910 g) was collected by filtration. The mother liquor was
concentrated, and a second crop of hydroxyphenol 3 was
obtained by recrystallizing the residue from hexanes-EtOAc
(60:40, by volume)15 to give a total of 1250 g of 3 (80%) as
white crystals; mp 106-107 °C (lit. mp 105-107 °C).10b,c
Purity (HPLC): 99.0%.
Conclusions
The tsetse fly attractant component 3-n-propylphenol (1)
has been prepared on a kilogram scale in two remarkable
simple steps in 75% overall yield. The procedure described
(11) As in ref 10c, our initial small-scale preparations of 3 employed diethyl
ether in which the starting aldehyde is more soluble than in THF although
solutions more dilute than the one described here were needed. However,
during the Mg-phenolate formation and subsequent Grignard reaction,
stirring became a serious problem. This solubility problem, exacerbated
by intensive cooling, should be the main reason for the earlier reported
low (58%) yield of 3: the Grignard adduct forms an ethyl ether-insoluble
double salt covering the surface of unreacted Mg-phenolate precipitate,
thus blocking complete consumption of the starting material. This could
also explain why even a large, 3.2-fold excess (see ref 10c) of EtMgBr
could not drive the reaction to completion. It is speculated that refluxing
the reaction mixture after the completion of the addition breaks up the
solid particles that include unreacted aldehyde phenolate.
(12) In preliminary experiments performed under various conditions on up to
50-g scales indicated (TLC) that the use of 2.2-2.4-fold excess of EtMgBr
led to intermediate 3 that was contaminated with some unreacted starting
material, the removal of which was cumbersome even by repeated
recrystallization (attempted distillation of the crude product led to degrada-
tion of 3). Furthermore, hydrogenation of the impure intermediate gave
the target phenol contaminated with m-cresol resulting from the reductive
deoxygenation of 2. Acceptable yield (80%) and excellent purity of 3 was
achieved when the excess of EtMgBr was increased to 2.6-fold, which is
significantly less than the 3.2 equiv used in ref 10c.
(13) Because continuous addition of the suspension of 2 to the Grignard reagent
presents some difficulties (clogging of the addition funnel), “inverse
addition” of EtMgBr solution to the vigorously stirred dispersion of the
aldehyde is preferred.
(14) Repeated extractions with 4 × 1 L methyl tert-butyl ether are necessary.
Measuring the volume of each extract indicated substantial amounts of
extractives present in the acidic aqueous phase: the volumes of the four
subsequent extracts were 2. 5, 2.2, 2.0, and 1.8 L, respectively.
(15) TLC analysis indicated that the mother liquor of the second crop contained
hydroxyphenol 3, some starting material, and other unidentified contami-
nants.
586
•
Vol. 7, No. 4, 2003 / Organic Process Research & Development